Skill documents are becoming the default adaptation surface for frozen large language model (LLM) agents, but current skill generators face two key limitations. First, humans and different LLMs differ in how they organize procedural knowledge and approach task solving. Skills authored from a human or single-model perspective may therefore fail to align with the reasoning behaviors of other backbones, limiting cross-model generalization. Second, sequential skill descriptions inadequately represent procedures with parallel branches and explicit dependencies, while existing skill optimizers primarily operate on prose and lack explicit support for graph-structured skill evolution. We introduce MultiEvoSkill, a skill-optimization framework comprising three core components: 1) Skill compiler casts the skill as a typed graph with tailored operators, producing a compact, informative representation with explicit procedural topology; 2) Heterogeneous agent system proposes revision advices using collective intelligence, where minimal and informative issues are identified based on multiple sets of experiences. 3) Structure-aware debate and evolution mechanism organizes candidate edits into a heterogeneous graph, resolves them under dependency and importance constraints, and distills the collective proposal into the artifact. Multi-dimensional gating ensures the information gain and topological correctness of each evolution. Optimal skills are then obtained when the system converges at a consistent state. Various experiments and ablations against SOTA methods validate the effectiveness and generalization of MultiEvoSkill.
Figures & tables
Figure 1: Illustration of two key advantages of our approach. (a) Graph-form skills explicitly encode procedural dependencies, branch conditions, and reusable actions, unlike linear prose. (b) Collective intelligence enables heterogeneous agents to exchange complementary experience and consolidate it into a single generalizable skill artifact.
Figure 2: Overview of MultiEvoSkill. Heterogeneous executors collect evidences for reflection; issue-specific proposer debates produce patches against one fixed parent. A hierarchical revisor merges patches, and the gate commits or rejects the resulting graph. Selection data enter the gate, not proposal generation.
Graph element
Procedural meaning
Decision 7
Decide whether results must remain formula-driven.
7true8
Create formulas with correct references when recalculation is required.
7false9
Compute and write materialized values when requested.
8→10 , 9→10
Continue to row merging after the selected branch.
Node 10 label
Preserve existing order; append only missing keys.
Edge note
Explain a guard or ordering rationale.
Table 1: Graph–document correspondence for the spreadsheet formula/value branch. Original node IDs are retained.
Method
WikiHow
ToolAlpaca
WebShop
SearchQA
SpreadsheetBench
Trace2Skill
53.04
35.30
28.69
86.88
50.00
CoEvoSkills
53.90
33.13
31.58
87.78
83.93
SkillOpt
53.39
53.54
34.54
88.14
82.14
SkillBoost
54.24
35.81
28.72
80.80
83.21
SkillGen
54.56
33.65
29.12
87.38
58.21
MultiEvoSkill
55.58
53.88
34.77
88.25
86.07
Table 2: Test-set scores across five benchmarks. Bold indicates the best result in a column.
Sample type
Required procedure
Why prose is fragile
Graph behavior
Lookup
Identify key row and target column; return exact cell value.
Aggregation and formatting rules are irrelevant but appear in the same prose skill.
Activate lookup branch only; skip aggregation and formatting subgraphs.
Formatting is a separate operation but prose may mix it with value computation.
Activate formatting branch with its own validation node.
Table 3: The illustrative prose skill combines several workflows in one paragraph, whereas a typed skill graph explicitly represents different branches for different samples while sharing common upstream nodes.
Variant
WikiHow
ToolAlpaca
WebShop
w/o δ1
45.88
27.71
31.53
w/o δ2
50.75
43.85
32.88
w/o δ3
50.75
45.69
33.03
w/o δ4
54.64
51.69
33.45
Full
55.58
53.88
34.77
Table 4: Proposer-panel ablations.
Variant
WikiHow
ToolAlpaca
WebShop
w/o g1
42.32
45.95
30.53
w/o g2
41.23
45.40
30.11
w/o g3
40.84
47.06
31.35
w/o g4
49.07
46.88
33.31
w/o g5
53.19
49.02
33.24
Full Gate
55.58
53.88
34.77
Table 5: Gating ablations.
Appendix figures & tables3 assets
Supplementary material from the paper’s appendix.
Appendix
Route
Representative methods
Trajectory distillation
Trace2Skill ( Ni et al., 2026 ) , SkillCAT ( Chen et al., 2026 ) , SkillEvolver ( Zhang et al., 2026a ) , Memento-Skills ( Zhou et al., 2026a ) , WikiSkill ( Tang et al., 2026 )
Execution-feedback refinement
SkillOpt ( Yang et al., 2026 ) , SkillBoost ( Lin et al., 2026 ) , SkillGrad ( Wang et al., 2026 ) , SkillRevise ( Liu et al., 2026 ) , EmbodiSkill ( Ju et al., 2026 )
Multi-agent evolution
CoEvoSkills ( Zhang et al., 2026b ) , SkillGen ( Ma et al., 2026 ) , MIND-Skill ( Li et al., 2026b ) , SkillHone ( Li and Hu, 2026 )
Population-based evolution
AgentPSO ( Hwang et al., 2026 ) , MOCHA ( Tanjim et al., 2026 ) , GEPA ( Agrawal et al., 2025 ) , EvoSkill ( Alzubi et al., 2026 )
Reinforcement-learning-based
D2Skill ( Tu et al., 2026 ) , SkillRL ( Xia et al., 2026a ) , SkillOS ( Ouyang et al., 2026 ) , Skill1 ( Shi et al., 2026b )
Appendix
Table 6: Major routes for agentic skill evolution.
Domain
n0
e
b
ϵ
τ
q0
WikiHow
40
4
40
.03
.01
.5
ToolAlpaca
40
4
40
.03
.01
.5
WebShop
40
4
40
.03
.01
.5
SpreadsheetBench
40
4
40
.10
.02
.5
SearchQA
40
4
40
.02
.01
.5
Appendix
Table 7: Evolution hyperparameters used in our experiments: initialization examples n0 , epochs e , batch size b , regression base tolerance ϵ , mean-gain threshold τ , and non-decrease fraction q0 .
Domain
Train
Selection
Test
WikiHow
157
53
52
ToolAlpaca
56
19
18
WebShop
80
27
26
SpreadsheetBench
80
40
280
SearchQA
400
200
1400
Appendix
Table 8: Dataset split sizes and evaluation protocol.
Skills can improve the performance of Large Language Model (LLM) agents by providing task-specific procedural guidance, while skill optimization further improves their effectiveness through iterative refinement. However, existing skill optimization methods typically represent skills as unstructured natural-language instructions, creating two key challenges: 1) Unstructured skills often lack explicit workflow-level guidance and contain substantial redundancy, making them difficult for LLMs to execute; 2) the vast search space of unconstrained natural-language skills makes skill optimization ineffective. To address these challenges, we propose representing skills as graph-structured natural-language artifacts. In graph-structured skills, each node represents an execution step together with its operational guidance, while directed edges encode context-dependent transitions between steps. Compared to unstructured skills, graph-structured skills can provide clear workflow-level guidance. Moreover, the proposed graph-structured skill can also facilitate skill optimization. Building on this structured representation, we introduce GraphSkillEvo, a population-based evolutionary optimization framework with mutation and crossover operators for graph-structured skills. By maintaining multiple candidate skills and combining effective components, GraphSkillEvo enables broader and more comprehensive exploration of the structured skill space than purely LLM-based iterative self-refinement. Extensive experiments across five agent benchmarks demonstrate that GraphSkillEvo consistently outperforms the strong skill optimization baseline SkillOpt, improving average accuracy by 4.01% on GPT-5.4-nano and 1.76% on GPT-5.4. Our code is available at https://github.com/ruisun7/GraphSkillEvo.
Rui Sun, Zhi Zheng, Zhenkun Wang +1
City University of Hong Kong · National University of Singapore · Southern University of Science and Technology
Natural-language skills are textual procedural memories through which large language model (LLM) agents retain reusable task knowledge without updating model weights. Existing methods typically treat skills as either static artifacts or monolithic documents optimized using aggregate validation scores as feedback. However, representing a skill as a monolithic document restricts optimization to its textual content, without explicitly modeling the structure through which procedural knowledge is retrieved and executed. We identify a key distinction between learning what knowledge to retain and determining how to organize it: textual updates should first be validated through execution evidence, after which the retained knowledge should be structured according to its procedural dependencies and retrieval requirements. To this end, we introduce SkillSpec, a two-phase framework comprising consensus-gated evolution and representation specialization. In the consensus-gated phase, complementary editing intents generate complete candidate skills. An update is committed only when paired evaluations reach consensus, requiring sufficient overall improvement and non-negative aggregate paired gain in every repeated evaluation. In the specialization phase, signals of process and redundancy sensitivity derived from the full optimization trajectory, including accepted and rejected candidates, guide the selection of a flat, graph, or hybrid representation.Across six benchmarks and three target language models, SkillSpec improves average success rate over SkillOpt by 6.89%, averaged across the three models. These results demonstrate that reliable skill evolution and representation specialization address complementary objectives: deciding what knowledge to retain and how to structure it for inference.
Anthropic proposes the concept of skills for LLM agents to tackle multi-step professional tasks that simple tool invocations cannot address. A tool is a single, self-contained function, whereas a skill is a structured bundle of interdependent multi-file artifacts. Currently, skill generation is not only label-intensive due to manual authoring, but also may suffer from human--machine cognitive misalignment, which can lead to degraded agent performance, as evidenced by evaluations on SkillsBench. Therefore, we aim to enable agents to autonomously generate skills. However, existing self-evolving methods designed for tools cannot be directly applied to skills due to their increased complexity. To address these issues, we propose CoEvoSkills, a self-evolving skills framework that enables agents to autonomously construct complex, multi-file skill packages. Specifically, CoEvoSkills couples a Skill Generator that iteratively refines skills with a Surrogate Verifier that co-evolves to provide informative and actionable feedback without access to ground-truth test content. On SkillsBench, CoEvoSkills outperforms five baselines on both Claude Code and Codex, and generalizes strongly to six additional LLMs. The code is publicly available at https://github.com/Zhang-Henry/CoEvoSkills.
Hanrong Zhang, Shicheng Fan, Henry Peng Zou +11
Steve · University of Illinois Chicago · MBZUAI +4